Rodent bedding, diet, and drinking water may contain compounds that affect the same endpoints that are the target of NTP's testing program, such as reproduction and development. For example, endotoxins (a byproduct of mold) can mask responses to exposure to asthma-causing agents. Likewise, phytoestrogens in the rodent diet can mask responses to exposure to estrogenic compounds. We are working with the NIEHS Quality Assurance Laboratory to quantify these effects and develop guidelines for limits on contaminants in bedding, diet and drinking water of rodent studies. We contributed to the study design of a multi-strain mouse study in which different strains are exposed to ionizing radiation then followed up for development of tumors. These strains were selected to represent a broad range of genetic variation. Although the data are not yet complete, the early indications are that different strains have a range of susceptibility to radiation and show varying locations and rates of tumors. Some compounds cause kidney tumors in rats exposed for two years. Two-year studies, however, are expensive and time-consuming. We have identified early onset kidney lesions that may predict later kidney tumors in rats. By evaluating data from a number of previous NTP studies, we confirmed the association between early onset lesions and kidney tumors, and showed that the association is not affected by type of diet or route of compound exposure. This finding suggests that short-term, less expensive studies may be useful in identifying kidney carcinogens.